TPS54302 circuitry for 12V output

[darkmatter_1729]

Hi ppl,
Can i use the below circuitry as 12V output, i tried simulating in proteus 8.11 but i think it is not supported so i didn't get output and Tina-Ti doesn't have the IC i'm using so if anybody can suggest as to where can i simulate any circuitry without platform hinderance or atleast simulate below circuit.
Thank you

 

[KlausST]

Hi,

Your circuit shows only minor differences to the datssheet schematic Fig. 8-1.
The differences are:
* additional input capacitor --> uncritical
* modified EN voltage --> uncritical
* modified feedback capacitor --> small change, uncritical
* modified feedback resistor (output voltage) --> uncritical
* three inductors in series. Functionally I see no problem. But due to lengthy wiring and bigger switching node and bigger enclosed switching loop area I expect a much higher EMI. So while the SMPS may work it still may fail compliance tests.

Also the PCB layout has big impact no only on EMI, but also on regulation stability and output noise.

Thus I see limited information you can gain from a simulation:
* it´s quite expectable that the circuit will work
* but you can not see the impact of the three inductors in series.
* you can not see the impact of your PCB layout.
* and you can not see EMI results.

And, btw: why three inductors in series at all? The absolutoe value is not that critical and thus you easily should be able to find a one inductor solution.

Klaus

 

[BradtheRad]

TI has a page for this device on its website.

TPS54302 data sheet, product information and support | TI.com

TI’s TPS54302 is a 4.5-V to 28-V Input, 3-A Output, EMI Friendly Synchronous Step-Down Converter. Find parameters, ordering and quality information

www.ti.com

They state you can use their design tools including a simulator PSPICE-for-TI.

A buck converter does not output a higher voltage than you input.
However the buck-boost type can accept your specified 8 to 28V, and output 12V inverted, if you can accept inverted polarity.

 

[darkmatter_1729]

Hi,

Your circuit shows only minor differences to the datssheet schematic Fig. 8-1.
The differences are:
* additional input capacitor --> uncritical
* modified EN voltage --> uncritical
* modified feedback capacitor --> small change, uncritical
* modified feedback resistor (output voltage) --> uncritical
* three inductors in series. Functionally I see no problem. But due to lengthy wiring and bigger switching node and bigger enclosed switching loop area I expect a much higher EMI. So while the SMPS may work it still may fail compliance tests.

Also the PCB layout has big impact no only on EMI, but also on regulation stability and output noise.

Thus I see limited information you can gain from a simulation:
* it´s quite expectable that the circuit will work
* but you can not see the impact of the three inductors in series.
* you can not see the impact of your PCB layout.
* and you can not see EMI results.

And, btw: why three inductors in series at all? The absolutoe value is not that critical and thus you easily should be able to find a one inductor solution.

Klaus

Thanks for detailed reply,

I had only 6.8 and 3.3 uH inductors and had to get an equivalent of anything near 15uH to get an output of 12V so chose this combination. So by your statement "but you can not see the impact of 3 inductors in series" does it mean its not possible to get 12V output?

Now with respect to the project i'm working on, i wanted to convert 16V supply to 12V using TPS54302 and the only purpose of this circuitry is to power arduino mega 2560, which will hardly take 1A of current. In that perspective can i achieve this using current circuitry or is to necessary to use a single 15uH inductor?

 

[darkmatter_1729]

TI has a page for this device on its website.

TPS54302 data sheet, product information and support | TI.com

TI’s TPS54302 is a 4.5-V to 28-V Input, 3-A Output, EMI Friendly Synchronous Step-Down Converter. Find parameters, ordering and quality information

www.ti.com

They state you can use their design tools including a simulator PSPICE-for-TI.

A buck converter does not output a higher voltage than you input.
However the buck-boost type can accept your specified 8 to 28V, and output 12V inverted, if you can accept inverted polarity.

Thank you for your reply,

I designed the circuit using same datasheet but the concern is about series combination of inductors, will it create any problems?

 

[KlausST]

Hi,

I had only 6.8 and 3.3 uH inductors

On a first test I´d use a single inductor with 6.8uH.
And if it does not work I´d use rather 2 x 6.8uH than 3 inductors.

Btw: the error to 15uH is less with 2x6.8uH than with your 3 inductor solution.
2x6.8uH gives just 10% error to 15uH. And often 15uH inductors have a tolerance of +/-20%.

***
Final circuit vs simulation:
I´m confused if you want to do the final circuit with one 15uH or several other inductors.
For me it´s more important to know how the final circuit performs than a ciruit with parts laying around.
What information do you gain on the 3 inductor solution vs a one inductor solution?

to get an output of 12V

The output voltage is primarily defined by the feedback, not by the inductors.

So by your statement "but you can not see the impact of 3 inductors in series" does it mean its not possible to get 12V output?

I clearly wrote that I "expect it to work" in the simulation.
But your "schematic only simulation" (without placement, signal routing, coupling effects) will not show how the circuit perform as a real circuit. Especially regariding loop stability and EMI.
So my doubt is: What information do you get from a simulation that most likely shows that a circuit (schematic) works, while it´s quite questionable whether the real circuit performs correctly.

to power arduino mega 2560

If you power the Arduino with 12V you waste 140% of energy (--> heat) compared to powering it with 5V.
What´s the benefit of powering it with 12V?

In that perspective can i achieve this using current circuitry or is to necessary to use a single 15uH inductor?

It depends on a lot of things: Mainly - as already written - on the PCB layout.
It it will be much harder to comply with EMI regulations when you use inductors in series.

I don´t know if yu are experienced with SMPS design and EMI compliant design. A good designer will have no troubles.
An amateur has way bigger troubles with 3 inductors instead of using just 1 inductor. SMPS design is not a simple task.

Did you consider to use a ready to buy module?

Klaus

Added: did you do the math given in the datasheet? You talk about 1A output current ... but use inductors with a saturation current of just 1A. Did you ignore the ripple current?

 

[darkmatter_1729]

Hi,

On a first test I´d use a single inductor with 6.8uH.
And if it does not work I´d use rather 2 x 6.8uH than 3 inductors.

Btw: the error to 15uH is less with 2x6.8uH than with your 3 inductor solution.
2x6.8uH gives just 10% error to 15uH. And often 15uH inductors have a tolerance of +/-20%.

***
Final circuit vs simulation:
I´m confused if you want to do the final circuit with one 15uH or several other inductors.
For me it´s more important to know how the final circuit performs than a ciruit with parts laying around.
What information do you gain on the 3 inductor solution vs a one inductor solution?


The output voltage is primarily defined by the feedback, not by the inductors.


I clearly wrote that I "expect it to work" in the simulation.
But your "schematic only simulation" (without placement, signal routing, coupling effects) will not show how the circuit perform as a real circuit. Especially regariding loop stability and EMI.
So my doubt is: What information do you get from a simulation that most likely shows that a circuit (schematic) works, while it´s quite questionable whether the real circuit performs correctly.


If you power the Arduino with 12V you waste 140% of energy (--> heat) compared to powering it with 5V.
What´s the benefit of powering it with 12V?


It depends on a lot of things: Mainly - as already written - on the PCB layout.
It it will be much harder to comply with EMI regulations when you use inductors in series.

I don´t know if yu are experienced with SMPS design and EMI compliant design. A good designer will have no troubles.
An amateur has way bigger troubles with 3 inductors instead of using just 1 inductor. SMPS design is not a simple task.

Did you consider to use a ready to buy module?

Klaus

Added: did you do the math given in the datasheet? You talk about 1A output current ... but use inductors with a saturation current of just 1A. Did you ignore the ripple current?

Thank you for your response sir,

This project was designed to make a in-house test bench for propuslion system and not be rolled out as a product. I got the point that using this combination of inductors will lead to EMI issues hence it has been decided that 15uH inductor will be used.

Regarding usage of 12V supply for arduino mega 2560, it has been mentioned in the article "https://docs.arduino.cc/learn/electronics/power-pins" that supply voltage for mega 2560 is between 7-12V. Since 12V is standard voltage it was decided that it will come handy if some extra circuitry wants 12V as input.

 

[KlausST]

Hi,

O.K. well done reading the document.
You are in the safe side if you keep on the document.

However, it says as long as you be careful, using the 5V input is O.K.
Precautions are
* do not connect USB at the same time when using external 5V supply. Because 5V may flow back to the computer when the external 5V is highervthan the USB_ 5V. This is explicitely the case when the compurt is powered down. The (back) current is not limited, thus may be destructive.
* do not connect anything to V_in. ... to avoid any current flow back from the (5V) voltage regulator.
* the same is true for the barrel jack

To protect against V_in and barrel jack problem there is a simple solution: connect a schottky diode from 5V (anode) to V_in (cathode). This is the usual method to prevent voltage regulator damage from reverse currents.

So it's fine to use the external power supply. If I had to use it, I'd rather go with the minimum input voltage plus headroom.
For your regulated power supply you don't need much headroom, since the voltage is reliable, predictable and stable.

The problem is the power dissipation.
With 8V input the power dissipation in the 5V regulator is 3W per ampere.
With 12V input the power dissipation in the 5V regulator is 7W per ampere.

Without heatsink the voltage regulator can stand 0.5W or maybe 1W. It is likely to overheat. (Sadly the document does not tell about this)
It is safe if you run the Arduino only from this power supply, since it will draw less than 100mA. But as soon as you connect external devices, like LED display or anything other that draws considerable current it may cause problems.

Klaus

 

[darkmatter_1729]

Hi,

O.K. well done reading the document.
You are in the safe side if you keep on the document.

However, it says as long as you be careful, using the 5V input is O.K.
Precautions are
* do not connect USB at the same time when using external 5V supply. Because 5V may flow back to the computer when the external 5V is highervthan the USB_ 5V. This is explicitely the case when the compurt is powered down. The (back) current is not limited, thus may be destructive.
* do not connect anything to V_in. ... to avoid any current flow back from the (5V) voltage regulator.
* the same is true for the barrel jack

To protect against V_in and barrel jack problem there is a simple solution: connect a schottky diode from 5V (anode) to V_in (cathode). This is the usual method to prevent voltage regulator damage from reverse currents.

So it's fine to use the external power supply. If I had to use it, I'd rather go with the minimum input voltage plus headroom.
For your regulated power supply you don't need much headroom, since the voltage is reliable, predictable and stable.

The problem is the power dissipation.
With 8V input the power dissipation in the 5V regulator is 3W per ampere.
With 12V input the power dissipation in the 5V regulator is 7W per ampere.

Without heatsink the voltage regulator can stand 0.5W or maybe 1W. It is likely to overheat. (Sadly the document does not tell about this)
It is safe if you run the Arduino only from this power supply, since it will draw less than 100mA. But as soon as you connect external devices, like LED display or anything other that draws considerable current it may cause problems.

Klaus

Ok will make sure that external devices consuming more than 100mA is not connected.

Thanks for support

 

[KlausST]

Hi,

Ok will make sure that external devices consuming more than 100mA is not connected.

Simple math:

100mA x ( 12V - 5V) = 700mW = 0.7W (heat disspated by the voltage regulator)
Now if there is an MC33269 in SOT223 package installed, then it´s thermal resistance
r_th_ja = 156°C/W (datasheet)
Temperatre rise is: 156°C/W x 0.7W = 105°C.(delta)
With an ambient temperature of 25°C you get a total temperature of 130°C.

If you build a case around your Arduino you get reduced air flow and thus even higher tempertures.

37°C is about human temperature
60°C is too hot to be touched with a finger
100°C is where water boils
130°C even is a lot higher
179°C is where Pb solder may melt.

With my applications I try to keep part temperatures below 60°C. (No need for others to do it the same way.)

For some electronic parts it is said that a temperature rise by 9°C reduces liftetime to 50%
18°C --> 25%
36°C --> 6%
105°C --> 0.03% (10 years will be reduced to 112 days)

Klaus